FAMILY. we-', <n a «tcn^ follows the order of thought of tiie speaker, so that it follows that the M'^'S think in an order of ideas different from that of the Khassis and the Mans. Own- to tiie existence of these differences we should not be justified in assuming a oommou oTicit- fol' fte -Won-Khmft languages on the one hand, and for the Munda, Kanoowiy, rod Malacca languages, on the other, We may, however, safely assume that thr*e is 'it. tlio bottom of all these tongues1 a common substratum, over which there have cettlcrl layers of A,lie speeches of other peoples, differing m different localities Neverlhc- Je«, tim &nbbtratum was firmly enough established to prevent its being entirely bidden by fiiera, and frequent, undeniable, traces of it are still discernible in languages spoken ia widely distant tracts of Nearer and Further India. Of what language this original substratum consisted, we are not yet in a position to gay. Whatever it was, it covered a wide area, larger than the area covered by many families of languages in India at the present day- Languages with this common substra- tum are now spoken not only in the modern Province of Assam, in Burma, Siam, Cambodia and Anam, but also over the whole of Central India as far west as the Berars, It is a far cry from Cochin China to Mmar, and yet, even at the present clay, the coin- cidences between the language of the Korkus of the latter District and the Anamese of Cochin China are strikingly obvious to any student of language who turns his attention to them. Still further food for reflection is given by the undoubted fact that, on the other sides the Miinda languages show clear traces of connexion with the speeches of the aborigines of Australia. This ancient substratum may have been the parent of the present Munda languages, or it may have been the parent of the present Mon-Khmer languages. It cannot have been the parent of both, but it is possible that it was the parent of neither, Logan, writing in the early fifties, believed that it is the Mon-Khmer family of which it was the parent, and that the speakers were a mixture of two distinct races, i.e., that Eastern Tibetans, or Western Chinese, came across the Himalaya, and mingled with the Australo- Dravidians of India proper, who are now looked upon as the aborigines of India, Porbes, in his Comparative Grammar, avoids the question, and contents himself with proving, what is now not a matter of doubt, that the Munda and Mon-Khmer families had no common parentage, Kuhn is more cautious than Logan. He proves the existence of the common substratum, but does not venture to state to what family of languages it belonged. Thomson does not deal with the question directly, but it may be gathered from the paper quoted below that his opinion is that most probably the substratum is a Munda one, and that a population akin to the Indian Munda races originally extended as far east as Further India. This was before the beginnings of those invasions from the north which resulted, first, in the Mon-Khmer, and, afterwards, in the Tibeto-Burman and Tai settle- ments in that region, AUTHORITIES-, The following writings deal with the general question of the Mon-Khmer races and languages:— r, J.R-/n»«n«(rf papers on the mno%, of *7,e Indo^Pacijlc Islands, winch appeared in the ,11LI MiPday°> tt*y a11 1* studied ™th advantage, though much has », Oangetic and Dravirian Languages, on pp. 186 and ff. of Vol. vii (1853). '---------------------,------------- 1 So Kuhn in ftc Bettrdge quoted below.